Recognition of noisy speech: using minimum-mean log-spectral distance estimation
HLT '90 Proceedings of the workshop on Speech and Natural Language
Acoustical and environmental robustness in automatic speech recognition
Acoustical and environmental robustness in automatic speech recognition
Reduced channel dependence for speech recognition
HLT '91 Proceedings of the workshop on Speech and Natural Language
Performance of SRI's DECIPHER™ speech recognition system on DARPA's CSR task
HLT '91 Proceedings of the workshop on Speech and Natural Language
Hi-index | 0.00 |
A new mapping algorithm for speech recognition relates the features of simultaneous recordings of clean and noisy speech. The model is a piecewise nonlinear transformation applied to the noisy speech feature. The transformation is a set of multidimensional linear least-squares filters whose outputs are combined using a conditional Gaussian model. The algorithm was tested using SRI's DECIPHER™ speech recognition system [1-5]. Experimental results show how the mapping is used to reduce recognition errors when the training and testing acoustic environments do not match.